Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
awslabs avatar

Hyperpod Version Checker

  • 117 installs
  • 850 repo stars
  • Updated August 3, 2026
  • awslabs/agent-plugins

hyperpod-version-checker is a Claude skill that checks and compares software component versions and compatibility across SageMaker HyperPod cluster nodes.

About

This skill checks and compares software component versions on SageMaker HyperPod cluster nodes, including NVIDIA drivers, CUDA, cuDNN, NCCL, EFA, MPI, Neuron SDK, Python, and PyTorch. It runs a bundled script that analyzes CUDA/driver compatibility and, with JSON output, makes cross-node diffing easy. A developer uses it to detect version mismatches, plan upgrades, or document cluster configuration.

  • Checks NVIDIA driver, CUDA, cuDNN, NCCL, EFA, OFI NCCL, MPI, Neuron SDK, Python, and PyTorch versions
  • Analyzes CUDA/driver compatibility and detects version mismatches across nodes
  • JSON output for easy multi-node diffing; uploaded and run via the hyperpod-ssm skill

Hyperpod Version Checker by the numbers

  • 117 all-time installs (skills.sh)
  • Ranked #541 of 1,039 Cloud & Infrastructure skills by installs in the Skillselion catalog
  • Data as of Aug 4, 2026 (Skillselion catalog sync)
At a glance

hyperpod-version-checker capabilities & compatibility

Capabilities
hyperpod ssm · hyperpod node debugger · hyperpod nccl · hyperpod performance debugger
Works with
aws
Use cases
devops · debugging · documentation
From the docs

What hyperpod-version-checker says it does

Check and compare software component versions on SageMaker HyperPod cluster nodes - NVIDIA drivers, CUDA toolkit, cuDNN, NCCL, EFA, AWS OFI NCCL, GDRCopy, MPI, Neuron SDK
SKILL.md
NCCL: Use 2.18+ for CUDA 12.x, 2.12+ for CUDA 11.x. Must be consistent across all nodes.
SKILL.md
npx skills add https://github.com/awslabs/agent-plugins --skill hyperpod-version-checker

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs117
repo stars850
Last updatedAugust 3, 2026
Repositoryawslabs/agent-plugins

What it does

Check and compare driver, CUDA, NCCL, and framework versions across HyperPod nodes for compatibility.

Who is it for?

Operators verifying driver and library versions or planning upgrades across HyperPod nodes

Skip if: Remediating faults, which the debugger skills handle

When should I use this skill?

Checking component versions, verifying compatibility, or detecting version drift across nodes

What you get

A per-node version report with CUDA/driver compatibility analysis, diffable across nodes.

  • Per-node component version report (text or JSON) with compatibility analysis

By the numbers

  • 11 components checked
  • 5 driver-series compatibility rows

Files

SKILL.mdMarkdownGitHub ↗

HyperPod Version Checker

Upload to cluster nodes via hyperpod-ssm skill, then execute.

Usage

# Text report to console + file
bash hyperpod_check_versions.sh

# JSON only to stdout (text report still saved to file) — best for piping/parsing
bash hyperpod_check_versions.sh --json

# Custom output file
bash hyperpod_check_versions.sh --output /tmp/versions.txt

# No color (for logging)
bash hyperpod_check_versions.sh --no-color

Output file: component_versions_<hostname>_<timestamp>.txt (default)

What It Checks

ComponentDetection MethodApplicable When
NVIDIA Drivernvidia-smiGPU instances (p3/p4/p5/g5)
CUDA Toolkitnvcc, /usr/local/cuda symlinkGPU instances
cuDNNHeader file, packagesGPU instances doing deep learning
NCCLLibrary filename, header, packagesDistributed GPU training
EFA/opt/amazon/efa_installed_packages, fi_infoEFA-capable instances (p4d/p4de/p5/trn1/trn2)
AWS OFI NCCLefa_installed_packages, library searchEFA + NCCL workloads
GDRCopyrpm/dpkg, kernel moduleGPU instances with RDMA (p4d+/p5)
MPImpirun, /opt/amazon/openmpiDistributed training
Neuron SDKneuronx-cc, neuron-ls, packagesTrainium/Inferentia (trn1/trn2/inf1/inf2)
Python/PyTorchpython3, torch importML workloads
Container runtimedocker, containerd, kubectl, nvidia-ctkEKS clusters

Multi-Node Comparison

Run on each node individually via the hyperpod-ssm skill. With --json, stdout is clean JSON for easy diffing.

Compatibility Reference

The script automatically analyzes CUDA/driver compatibility. For reference:

Driver SeriesSupported CUDA
580+13.x, 12.x, 11.x
570+12.8+ (Blackwell), 12.x, 11.x
545+12.3-12.7, 11.x
525-53512.0-12.2, 11.x
450+11.x only

NCCL: Use 2.18+ for CUDA 12.x, 2.12+ for CUDA 11.x. Must be consistent across all nodes.

EFA InstallerAWS OFI NCCL
1.29+v1.7.3+ (recommended)
1.26-1.28v1.7.0-v1.7.2
1.20-1.25v1.6.0+

Related skills

FAQ

How do I compare versions across nodes?

Run the script on each node via the hyperpod-ssm skill; with --json, stdout is clean JSON for easy diffing.

Does it check compatibility?

Yes. It automatically analyzes CUDA/driver compatibility and includes a reference table of driver series to supported CUDA versions.

Cloud & Infrastructureinframonitoring

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.